1. Field of the Invention
The present invention relates, in general, to microspheres and, more specifically, to methods of manufacturing microspheres.
2. Description of the Art
Methods to produce hollow microspheres potentially suitable for insulation materials have been disclosed in U.S. Pat. Nos. 3,030,215; 3,161,463; 3,365,315; 3,888,957 and 4,012,290. With few exceptions, notably, U.S. Pat. No. 4,349,456, all of the current methods to produce hollow microspheres or shells rely on the use of a blowing gas, i.e., a gas which evolves within a drop of melted glass and blows the drop into a hollow glass bubble or sphere.
The input or feed particles in these processes are either drops formed from aqueous glass solutions or solid "frit" particles which become fluid upon sufficient heating. Specifically, the frit particles are generally heated to a temperature between approximately 1,000.degree. C. to 1,800.degree. C. In those processes using small, solid frit particles, the shells are formed by introducing the particles into a heated zone, i.e., into a furnace or torch flame, to thereby raise the particles' temperatures to the range in which the glass exhibits the properties of the liquid. Specifically, the particles are heated to the temperature range in which the viscosity of the glass is sufficiently low so that the glass flows readily, i.e., the glass viscosity is less than 500 poise and, preferably, less than 100 poise.
When a sufficiently low viscosity has been attained, the surface tension of the glass, i.e., 100 to 400 dynes/cm at the specified temperature, causes the formation of spheroidal glass drops. Within these drops, microscopically small bubbles are formed by gases which are generated by the volatilization of blowing agents which have been incorporated into the glass feed or frit particle. These smaller bubbles coalesce to form a single void within the glass drop thereby producing a glass bubble.
In the majority of commercial processes used today, the gases are generated by incorporating into the frit particles various blowing agents, i.e., materials which upon heating will evolve gases. While in the heated zone, the glass bubbles expand from the blowing gases generated by the blowing agents and from the in-permeation of ambient gases. When the bubbles exit the heated zone and are exposed to normal room temperatures, because of their small heat capacity, the bubbles rapidly solidify thereby forming solid glass shells. The internal pressure of the bubbles at the point of exiting the heated region is balanced by the forces resulting from the combination of the surface tension of the glass and the external ambient pressure. The resulting shells have diameters ranging from about 5 .mu.m to approximately 5,000 .mu.m. The distribution of diameters of the resulting shells depends on the production method used, the size of the feed particles, the temperature history during the blowing process, the composition of the glass, and the type of ambient gases.
In all of the current commercial shell or microsphere manufacturing processes, the shells contain residual gases. These residual gases are captured during the blowing process and, as indicated above, are a combination of gases resulting from the volatilization of the blowing agents and from the influx of ambient gases. The present invention relates to a method by which these residual gases can be removed as their presence increases the heat transfer characteristics of the hollow shells. The present method is based on the out-permeation of the residual gases into controlled environments which leads to a substantial evacuation of the shells so as to obtain the maximum possible insulation value.
A successful out-permeation of residual gases from borosilicate glass shells, type B12AX produced by the 3M Corporation, was reported by Parmley and Cunnington at "An Ultralightweight, Evacuated, Load-Bearing, High-Performance Insulation System", proceedings of the 2nd AIAA and ASME Thermophysics and Heat Transfer Conference, Palo Alto, Calif., 1978. The residual gases in these shells were determined by the authors to be 99.97% SO.sub.2 and 0.03% air, at a total pressure of 2.1.times.10.sup.2 Torr. According to the authors, by baking the shells in vacuum at 421.degree. C. for twenty days, internal pressures of .ltoreq.0.1 mTorr, which is the requisite pressure for good thermal insulation values, were obtained. At this pressure, the mean free path of the gas molecules is large compared to the dimensions of the shell and the conduction through the gas is linear with the pressure and becomes negligibly small. Although Parmley and Cunnington reported that baking the shells caused the SO.sub.2 to permeate the walls of the sodium borosilicate glass shells, it is also possible that the SO.sub.2 was caused to react with the sodium in the walls of the shells forming a low vapor pressure solid, such as NA.sub.2 SO.sub.4. Alternately, due to the relatively high solubility of SO.sub.2 in glass, maintaining the shells at the elevated temperatures may have caused the SO.sub.2 to be re-dissolved into the walls of the shells.
Tests conducted by the inventor indicate that a reduction in SO.sub.2 is indeed accomplished according to the method described by Parmley and Cunnington. However, the final pressure is approximately 6 Torr of oxygen, far from the required pressure of 0.1 mTorr, with the pressure of the oxygen remaining essentially unchanged for baking periods of approximately two months. In addition, the residual gas mixture was not that reported by Parmley and Cunnington, but was determined to be one-third oxygen and two-thirds SO.sub.2. The causes for the difference in the results are unknown, but it is clear that this procedure will not provide vacuum shells with commercially available materials because of the impractically slow out-permeation of oxygen from the shells.
Torobin, in U.S. Pat. No. 4,303,732, reveals a process for manufacturing vacuum microspheres. This process is based on blowing individual shells from a molten glass using a coaxial nozzle. The blowing gas flows through the inner nozzle and the glass flows through the annulus between the two coaxial nozzles thereby forming a glass bubble, which is subsequently detached by vibration or the action of a second transverse gas stream. The vacuum is formed by entraining into the blowing gas particles of metal or metal organic compounds which will become gaseous at the molten glass temperature and which, upon cooling, will re-solidify and form a reflective film on the inner surface of the microshell. The residual gas pressure at room temperature is that due to the vapor pressure of the metal. With a proper choice of metals, the film will exhibit the requisite low vapor pressure, i.e., less than 0.1 mTorr. Apparently, this process has been successfully demonstrated. However, it has also apparently proved too costly to be commercially viable.
Coxe, in U.S. Pat. No. 3,607,169, discloses a similar process. Again the production of the shell is based on the use of metals as blowing agents. In this case, glass-coated metal particles are made by an extrusion method. This process is also based on the encapsulating material being pumped through the annulus of a coaxial nozzle. As in the case of the Torbin patent, described above, this process, although apparently successfully demonstrated, has proven too costly to scale up to production levels. In general, droplet generation appears unsuitable for high volume production of evacuated microshells.
Sowman, in U.S. Pat. No. 4,349,456, discloses a process for making ceramic metal oxide shells which is not based on encapsulating blowing agents, i.e., the process does not utilize a blowing gas. This process consists of forming drops of an aqueous metal oxide colloidal sol in a dehydrating liquid, which rapidly removes the water from the drops thereby forming a gelled microcapsule. These microcapsules are recovered by filtration, dried and fired to convert them into shells. Prior to firing, the microcapsules are sufficiently porous that, if placed in a vacuum during the firing process, the gases can be removed and the resulting shells will generally be impermeable to ambient gases. However, this process is also not cost effective in scaling up to high volume production levels required for insulation because of high material costs, and the costs of purchasing and maintaining vacuum systems.
Thus, it would be desirable to provide a hollow microsphere or shell for use in thermal insulating material which has a substantially evacuated interior. It would also be desirable to provide a microsphere or shell which can be produced in quantity by a cost effective method. It would also be desirable to provide a method of manufacturing microspheres or shells having evacuated interiors with pressures less than 0.1 mTorr.